Color perception is part of the fabric of human experience. Homer (c. 1100 b. c.) writes of “the rosy-fingered dawn”. Lady Murasaki no Shikibu (c. 1000 a.d.) uses word colors (“purple, yellow shimmer of dresses, blue paper”) in the world's first novel. In the early nineteenth century Thomas Young, an English physician, proposed a trichromatic theory of color vision, based on the action of three different retinal receptors. Fifty years later James Clerk Maxwell, the British physicist and Hermann von Helmholtz, the German physiologist, independently showed that all of the colors we see can be made up from three suitable spectral color lights. In 1964 Edward MacNichol and colleagues at Johns Hopkins and George Wald at Harvard measured the absorption by the visual pigments in cones, which are the color receptor cells. Rods are another type of photoreceptor cell in the primate retina. These cells are more sensitive to dimmer light but are not directly involved in color perception. The individual cones have one of three types of visual pigment. The first is most sensitive to short waves, like blue. The second pigment is most sensitive to middle wavelengths, like green. The third pigment is most sensitive to longer wavelengths, like red.
The retina can be thought of a big flower on a stalk where the top of that stalk is bent over so that the back of the flower faces the sun. In place of the sun think of the external light focused by the lens of the eye onto the back of the flower. The cones and rods cells are on the front of the flower; they get the light which has passed through from the back of the somewhat transparent flower. The photoreceptor nerve cells are connected by synapses to bipolar nerve cells, which are then connected to the ganglion nerve cells. The ganglion nerve cells connect to the optic nerve fibers, which is the “stalk” that carries the information generated in the retina to the brain. Another type of retinal nerve cell, the horizontal cell, facilitates the transfer of information horizontally across bipolar cells. Similarly, another type of cell, the amacrine facilitates the horizontal transfer of information across the ganglion cells. The interactions among the retinal cells can be quite complex. On-center and off-center bipolar cells can be stimulated at the same time by the same cone transmitter release to depolarize and hyperpolarize, respectively. A particular cell's receptive field is that part of the retina, which when stimulated, will result in that cell's stimulation. Thus, most ganglion cells would have a larger receptive field than most bipolar cells. Where the response to the direct light on the center of a ganglion cells receptive field is antagonized by direct light on the surround of its receptive field, the effect is called center-surround antagonism. This phenomenon is important for detecting borders independent of the level of illumination. The existence of this mechanism for sharpening contrast was first suggested by the physicist Ernst Mach in the late 1800's. More detailed theories of color vision incorporate color opponent cells. On the cone level, trichromatic activity of the cone cells occurs. At the bipolar cell level, green-red opponent and blue-yellow opponent processing systems of the center-surround type, occur. For example, a cell with a green responding center would have a annular surround area, which responded in an inhibiting way to red. Similarly there can be red-center responding, green-surround inhibiting response. The other combinations involve blue and yellow in an analogous manner.
It is widely known that Galvani, around 1780, stimulated nerve and muscle response electrically by applying a voltage on a dead frog's nerve. Less well known is that in 1755 LeRoy discharged a Leyden jar, i.e., a capacitor, through the eye of a man who had been blinded by the growth of a cataract. The patient saw “flames passing rapidly downward.”
In 1958, Tassicker was issued a patent for a retinal prosthetic utilizing photosensitive material to be implanted subretinally. In the case of damage to retinal photoreceptor cells that affected vision, the idea was to electrically stimulate undamaged retinal cells. The photosensitive material would convert the incoming light into an electrical current, which would stimulate nearby undamaged cells. This would result in some kind of replacement of the vision lost. Tassicker reports an actual trial of his device in a human eye. (U.S. Pat. No. 2,760,483).
Subsequently, Michelson (U.S. Pat. No. 4,628,933), Chow (U.S. Pat. Nos. 5,016,633; 5,397,350; 5,556,423), and De Juan (U.S. Pat. No. 5,109,844) all were issued patents relating to a device for stimulating undamaged retinal cells. Chow and Michelson made use of photodiodes and electrodes. The photodiode was excited by incoming photons and produced a current at the electrode.
Normann et al. (U.S. Pat. No. 5,215,088) discloses long electrodes 1000 to 1500 microns long designed to be implanted into the brain cortex. These spire-shaped electrodes were formed of a semiconductor material.
Najafi, et al., (U.S. Pat. No. 5,314,458), disclosed an implantable silicon-substrate based microstimulator with an external device which could send power and signal to the implanted unit by RF means. The incoming RF signal could be decoded and the incoming RF power could be rectified and used to run the electronics.
Difficulties can arise if the photoreceptors, the electronics, and the electrodes all tend to be mounted at one place. One issue is the availability of sufficient area to accommodate all of the devices, and another issue is the amount of power dissipation near the sensitive retinal cells. Since these devices are designed to be implanted into the eye, this potential overheating effect is a serious consideration.
Since these devices are implants in the eye, a serious problem is how to hermetically seal these implanted units. Of further concern is the optimal shape for the electrodes and for the insulators, which surround them. In one embodiment there is a definite need that the retinal device and its electrodes conform to the shape of the retinal curvature and at the same time do not damage the retinal cells or membranes.
The length and structure of electrodes must be suitable for application to the retina, which averages about 200 microns in thickness. Based on this average retinal thickness of 200 microns, elongated electrodes in the range of 100 to 500 microns appear to be suitable. These elongated electrodes reach toward the cells to be activated. Being closer to the targeted cell, they require less current to activate it.
In order not to damage the eye tissue there is a need to maintain an average charge neutrality and to avoid introducing toxic or damaging effects from the prosthesis.
A desirable property of a retinal prosthetic system is making it possible for a physician to make adjustments on an on-going basis from outside the eye. One way of doing this would have a physician's control unit, which would enable the physician to make adjustments and monitor the eye condition. An additional advantageous feature would enable the physician to perform these functions at a remote location, e.g., from his office. This would allow one physician to remotely monitor a number of patients remotely without the necessity of the patient coming to the office. A patient could be traveling distantly and obtain physician monitoring and control of the retinal color prosthetic parameters.
Another version of the physician's control unit is a hand-held, palm-size unit. This unit will have some, but not all of the functionality of the physician's control unit. It is for the physician to carry on his rounds at the hospital, for example, to check on post-operative retinal-prosthesis implant patients. Its extreme portability makes other situational uses possible, too, as a practical matter.
The patient will want to control certain aspects of the visual image from the retinal prosthesis system, in particular, image brightness. Consequently, a patient controller, performing fewer functions than the physician's controller is included as part of the retinal prosthetic system. It will control, at a minimum, bright image, and it will control this image brightness in a continuous fashion. The image brightness may be increased or decreased by the patient at any time, under normal circumstances.
A system of these components would itself constitute part of a visual prosthetic to form images in real time within the eye of a person with a damaged retina. In the process of giving back sight to those who are unable to see, it would be advantageous to supply artificial colors in this process of reconstructing sight so that the patient would be able to enjoy a much fuller version of the visual world.
In dealing with externally mounted or externally placed means for capturing image and transmitting it by electronic means or other into the eye, one must deal with the problem of stabilization of the image. For example, a head-mounted camera would not follow the eye movement. It is desirable to track the eye movements relative to the head and use this as a method or approach to solving the image stabilization problem.
By having a method and apparatus for the physician and the technician to initially set up and measure the internal activities and adjust these, the patient's needs can be better accommodated. The opportunity exists to measure internal activity and to allow the physician, using his judgment, to adjust settings and controls on the electrodes. Even the individual electrodes would be adjusted by way of the electronics controlling them. By having this done remotely, by remote means either by telephone or by the Internet or other such, it is clear that a physician would have the capability to intervene and make adjustment as necessary in a convenient and inexpensive fashion, to serve many patients.